Introduction
Waḍʿ (from وضع, waḍaʿa, yaḍaʿu: “to put down”) denotes the coinage of a word. Stemming from a root with a very general meaning, its exact use as a technical term may seem difficult to pinpoint. The Qurʾan invokes the “proper place” of a word (mawḍiʿ, pl. mawāḍiʿ), a derivative of waḍaʿa, in the context of the distortion of scripture: “they bend the words away from their places (mawāḍiʿhī)” (Q 4/al-Nisāʾ: 46, Q 5/al-Māʾida: 13; 41). In treatises on solecism (laḥn al-ʿāmma), a similar phrase is employed for the incorrect use of a word: “what they [i.e., language users] put down in a place other than the word’s own place” (mā waḍaʿūhu ghayra mawḍiʿihī).1E.g., in Ibn Makkī al-Ṣaqalī, Tathqīf al-lisān (ed. Matar), 197–207. In ḥadīth scholarship, mawḍūʿ, the passive participle, is used to indicate a fabricated piece from prophetic tradition (ḥadīth), a ḥadīth that was “put in place.”
In linguistic contexts, waḍʿ is used in discussions of the origin of language and thus touches on core questions of language philosophy: how and by whom was language established and what is the relation between vocal form (lafẓ) and meaning (maʿnā)? Arabic language theorists discussed whether language was a matter of divine inspiration—God assigning vocal form to meaning—or convention (tawāḍuʿ, iṣṭilāḥ), an agreement between humans. “Coinage” in this context is a powerful translation for waḍʿ as it evokes the minting of a coin, the act of stamping it with an image.
Originating as a philosophical and theological question,2Still relevant is Bernard Weiss’s unpublished PhD dissertation “Language in Orthodox Muslim Thought”: A Study of ‘Wad‘ Al-Lughah’ and Its Development;” he continued his work on waḍʿ in several articles on the subject and in the entry ʿilm al-waḍʿ in the Encyclopedia of Arabic Language and Linguistics, see Weiss, “Medieval Muslim Discussions of the Origin of Language”; Weiss, “ʿIlm al-waḍʿ”; Weiss, “Waḍʿ al-luġa”. For use of the term waḍʿ in grammatical theory, see Sheyhatovitch, The Distinctive Terminology in Šarḥ al-Kāfiya by Raḍī al-Dīn al-ʾAstarābāḏī, 75–127. In his unpublished PhD dissertation, Giovanni M. Carrera discussed the development of ʿilm al-waḍʿ as a discipline from the eighth/fourteenth century up until today, building on the foundation laid by Weiss: “The Development of ʿilm al-waḍʿ”. Also see Carrera and Chiabotti, “Origine et finalité du langage dans le moyen âge islamique.” the discussion on waḍʿ evolved into a theory of semantics and pragmatics, called the discipline of waḍʿ (ʿilm al-waḍʿ), that became increasingly comprehensive between the eighth/fourteenth and fourteenth/twentieth century. In modern scholarship, al-Risāla al-waḍʿiyya, a two-folio treatise devoted to the details of waḍʿ by the theologian ʿAḍud al-Dīn al-Ījī (d. 756/1355), has been viewed as the starting point of this discipline.3Modern editions mostly depart from a sharḥ or ḥāshiya on al-Risāla al-Waḍʿiyya. See, e.g., al-Qūshjī, Sharḥ al-risāla al-ʿaḍudiyya (ed. al-Fathi). The text of the Waḍʿiyya itself can be found in collections of mutūn, short prose texts and poems meant for memorization. See, e.g., Marʿī, al-Majmūʿ al-kabīr, where it is grouped with texts on rhetoric. Al-Ījī’s dense treatment of the modes of signification ascribed to certain parts of speech indeed gave the impulse for commentaries and handbooks that elaborated upon the text and expanded the theory, triggering its development into a teaching subject and, ultimately, its integration into classifications of the sciences. However, here I wish to focus not on this formalized body of literature prompted by al-Ījī’s Risāla, but on the broader linguistic notion of waḍʿ we find in language scholarship and jurisprudence prior to the eighth/fourteenth century.
Corpus
Against the backdrop of the language-philosophical foundation of waḍʿ as discussed by the fourth/tenth-century language theorists Ibn Jinnī (d. 392/1002) and Ibn Fāris (d. 395/1004), I will highlight the role of waḍʿ in uṣūl al-fiqh. Implicit in Muḥammad b. Idrīs al-Shāfiʿī’s (d. 204/820) al-Risāla, the concept was put to different uses in Kitāb al-Muʿtamad fī uṣūl al-fiqh by the Muʿtazilī jurist Abū l-Ḥusayn al-Baṣrī (d. 436/1044) and in the works of the influential uṣūlīs Imām al-Ḥaramayn al-Juwaynī (d. 478/1084) and Abū Bakr al-Bāqillānī (d. 403/1013). My discussion of the use of waḍʿ by these scholars serves as an illustration and point of reference and by no means as an exhaustive treatment of waḍʿ in Islamic legal theory. I use printed editions for reference. While I briefly refer to scholars after al-Ījī such as al-Sharīf al-Jurjānī (d. 816/1413) and ʿAlī al-Qushjī (d. 879/1474) to point to the further systematic development of ʿilm al-waḍʿ in the conclusion, I will not discuss the summarizing treatment of waḍʿ by language scholars of the post-formative period such as Jalāl al-Dīn al-Suyūṭī (d. 911/1505).
Terminological Analysis and Historical Semantics
Waḍʿ and the Origin of Language
Ibn Jinnī and Ibn Fāris offered different answers to the question of the origin of language (aṣl al-lugha). Ibn Fāris asserts that language is a matter of divine instruction (tawqīf) and not of convention (iṣṭilāḥ).4For historical accounts of the tawqīf/iṣṭilāḥ-debate and its Muʿtazilī background, see Versteegh, “Linguistic Attitudes and the Origin of Speech in the Arab World” and Shah, “Classical Islamic Discourse on the Origins of Language.” He concurs with the Companion Ibn ʿAbbās, who suggested the following interpretation of the Qurʾanic statement (Q 2/al-Baqara: 31) “and He taught Adam all the names/words”: “He taught Adam all the names, and they are those [names] the people learn among themselves: of the animal, land, plain, mountain, donkey, and the like of the peoples and other things.” Proof (dalīl) of this, according to Ibn Fāris, is the fact that scholars use kalām al-ʿarab—pure Bedouin speech as reflected in pre- and early-Islamic poetry and prose—as a basis for the interpretation of scripture: “If language were a matter of convention (muwāḍaʿa wa-ṣṭilāḥ), they would not rather use them (the pre- and early-Islamic Bedouin) for corroboration (iḥtijāj, “providing proof”) than us.”5Ibn Fāris, al-Ṣāḥibī fī fiqh al-lugha al-ʿarabiyya (ed. Saqr), 4. For a recent discussion of the question of the origin of language, see Amharar, “L’origine du langage en islam”Specifically on the views of Ibn Fāris and Ibn Jinnī, see Larcher, “‹Et Allah apprit à Adam tous les noms…›”
Ibn Jinnī concludes in al-Khaṣāʾiṣ that the question whether language is a matter of revelation/divine inspiration or convention (iṣṭilāḥ, tawāḍuʿ) must remain unsolved (waqf), even though most theorists opt for the conventionalist position.6Ibn Jinnī himself, however, prefers the view that language derives from onomatopoeia (1: 47), which is a “naturalist” rather than a “revelationist” position, but closely connected to the revelationist position in that God is ultimately the creator of these “natural” sounds. See also Weiss, “Medieval Muslim Discussions of the Origin of Language”, for an account of the naturalist, revelationist, and conventionalist positions in disciplines other than lugha. The concomitant question of who the wāḍiʿ, “establisher” of language, was—perhaps a group of early sages—is not deemed important. Ibn Jinnī describes the process of iṣṭilāḥ or tawāḍuʿ as a practical way of expressing things that are beyond the range of vision.7Ibn Jinnī, al-Khaṣāʾiṣ (ed. al-Najjar), 1, 44–45:
وذلك أنهم ذهبوا إلى أن أصل اللغة لا بد فيه من المواضعة قالوا: وذلك كأن يجتمع حكيمان أو ثلاثة فصاعداً فيحتاجوا إلى الإبانة عن الأشياء المعلومات فيضعوا لكل واحد منها سمة ولفظاً إذا ذكر عرف به ما مسماه، ليمتاز من غيره وليغنى بذكره عن إحضاره إلى مرآة العين …
This theoretical debate had little consequence for the practice of language scholarship. Most lexicographers contented themselves with collecting the Arabic lexicon and citing textual witnesses to confirm the meaning and use of a word in Qurʾan, ḥadīth, and kalām al-ʿarab (Bedouin speech). A notable exception was Ibn Fāris, who tried to identify a common “basic meaning” (aṣl) of verbal roots in his dictionary. He explains that there are two types of words in kalām al-ʿarab: those subject to qiyās and “what was coined through waḍʿ.”8Ibn Fāris, Muʿjam maqāyīs al-lugha (ed. Haroun), s.v. al-qarr:
وقد قلنا إنَّ كلامَ العرب ضربان: منه ما هو قياسٌ، وقد ذكرناه، ومنه ما وُضِع وضعاً، وقد أثبَتْنا ذلك كلَّه. واللّه أعلم.
The latter expression is used in the dictionary to account for a minority of cases in which there is no qiyās, i.e., no identifiable rationale for the meaning of a root.9Maqāyīs al-lugha contains fewer than thirty roots that are described as mā wuḍiʿa waḍʿan; they mainly concern quadri- and quinqueliterals. Thus, waḍʿ denotes the assignment of vocal form to meaning by convention (iṣṭilāḥ).
Waḍʿ in Legal Theory
Waḍʿ and derived forms (mostly muwāḍaʿa, tawāḍuʿ, forms connoting mutual agreement) are used by lexicographers and grammarians of the formative period of Arabic scholarship to discuss the question of the origin of language. However, apart from Ibn Fāris in Maqāyīs al-lugha, they do not operationalize the concept of waḍʿ to discuss the meaning or etymology of individual roots or lexemes.
In uṣūl al-fiqh, the theoretical discussion of the origin of language retreats to the background. Here, waḍʿ is a productive concept and an epistemological category. The term’s implications are most readily visible when it is contrasted with another term. I will illustrate this through the dichotomies waḍʿ/majāz and waḍʿī/ʿaqlī, before outlining the further development of waḍʿ as a theory of semantics and pragmatics.
Ambiguity of the Arabic Language
Islamic legal scholars acknowledge the centrality of language for jurisprudence. The earliest treatment of legal hermeneutics (uṣūl al-fiqh), al-Shāfiʿī’s al-Risāla, stresses the necessity of proficiency in the Arabic language for the interpretation of the law, and highlights the semantic features of Arabic that complicate interpretation: God addresses people with statements “of which the sense seems unrestricted (ʿāmm) and readily apparent (ẓāhir)” and which “are in fact intended as unrestricted,” but also with statements of which the sense seems unrestricted but which are in fact partially restricted, or entirely restricted (khāṣṣ);10al-Shāfiʿī, al-Risāla (ed. Shakir), 52–53. My translation of the terms is loosely based on Shāfiʿī, The Epistle on Legal Theory: A Translation of al-Shāfiʿī’s Risālah, 27–28. or, in other words:11al-Shāfiʿī, al-Risāla (ed. Shakir), 222.
وكان اللِّسانُ كما وصفْتُ قبل هذا مُحْتَمِلاً للمعاني وأن يكون كتابُ الله يَنْزِلُ عامًّا يُرَادُ به الخاصَّ وخاصاً يراد به العام وفرضاً جملةً بيّنه رسول الله.
As I have shown earlier, language is susceptible to more than one meaning, and the Book of God sends down general statements that are meant to be taken as restricted, and restricted statements that are meant to be taken as unrestricted, and general obligations which the Messenger of God has explained.
As we will see below, differentiating between unrestricted (ʿāmm) and restricted (khāṣṣ) would prove to be one of the main concerns ofʿilm al-waḍʿ when it emerged as a discipline. The author’s point here is that Qurʾanic texts can be disambiguated with reference to the prophetic tradition (ḥadīth), a method that was not yet universally acknowledged in his time. Whereas al-Shāfiʿī does not use the word waḍʿ in a technical sense in al–Risāla, we find it employed in later works on uṣūl al-fiqh to highlight the problem of linguistic ambiguity as reflected, for instance, in literal and figurative language.
Waḍʿ and majāz
Abū al-Ḥusayn al-Baṣrī uses several forms of the root waḍaʿa to discuss the relation of literal and figurative usage in Kitāb al-Muʿtamad fī uṣūl al-fiqh:12Al-Baṣrī, al-Muʿtamad (ed. Hamidullah), 1: 995.
الْحَقِيقَة اللُّغَوِيَّة هِيَ مَا أفيد بِهِ مَا وضع لَهُ فِي أصل اللُّغَة […] وَاعْلَم أَنه إِذا حد الْمجَاز بِأَنَّهُ مَا أفيد بِهِ معنى مصطلحا عَلَيْهِ غير الْمَعْنى المصطلح عَلَيْهِ فِي الأَصْل بَطل إِذا وضع أهل اللُّغَة اسما لشَيْء ثمَّ تواضعوا على أَن يَجْعَلُوهُ اسما لشَيْء آخر وَلم ينقلوه عَن الأول لِأَنَّهُ قد أفيد بِهِ غير مَا وضع لَهُ فِي الأَصْل وَهُوَ مَعَ ذَلِك حَقِيقَة فِيهِ من جِهَة اللُّغَة كَمَا أَنه حَقِيقَة فِي الأول.
The linguistically “literal” is what conveys the sense for which it was coined originally in language […] Know that if “figurative” is defined as what conveys the sense that was convened upon and not the sense that was originally convened upon, this is void if the language scholars coined a name for something, then established that they make it a name for something else, and did not transfer it from the first [i.e., rescind the first sense], because it conveys a sense other than what it was coined for originally and it still has the literal sense from the viewpoint of language as it has in its first sense.
Abū l-Ḥusayn al-Baṣrī is reflecting a current debate on the relation of the literal and the figurative. Some scholars claimed that there were two unrelated instances of waḍʿ: the original waḍʿ and muwāḍaʿa (or tawāḍuʿ), the establishment of a secondary sense. Abū l-Ḥusayn, on the contrary, maintains that there must be a connection between both meanings, otherwise a metaphor would not make sense (the classic example is asad: a lion and a courageous man).13Cf. Weiss, “Language in Orthodox Muslim Thought,” 50–54. We note that the term waḍʿ is now used for the initial coinage of a word. Abū l-Ḥusayn al-Baṣrī consciously differentiates between waḍʿ (fī aṣl al-lugha) and muwāḍaʿa to make his point; other scholars circumscribed the secondary meaning (e.g., with baʿd al-muwāḍaʿa, where muwāḍaʿa refers to the initial coinage).
In a section on the parts of speech (aqsām al-kalām) in his abridgement (mukhtaṣar) al-Waraqāt, Imām al-Ḥaramayn al-Juwaynī (d. 478/1084) mentions three possible divisions of speech: first, the grammarians’ division into ism, fiʿl, and ḥarf (noun, verb, and particle); second, the legally relevant division into semantic functions: amr, nahy, khabar, istikhbār (command, prohibition, statement, question); third, a hermeneutic division into ḥaqīqa and majāz, i.e., literal and figurative language:14Al-Juwaynī, al-Waraqāt fī uṣūl al-fiqh (ed. Abd), 11. Similar definitions are omnipresent in the fifth/eleventh century, e.g., in al-Rāghib al-Iṣfahānī (d. 502/1108). See Key, Language between God and the Poet, 101ff. and notes.
فالحقيقة: ما بقي في الاستعمال على موضوعه، وقيل: ما استعمل في ما اصطلح عليه من المخاطبة.
والمجاز: ما تجوّز عن موضوعه.
And “literal” is what remains in usage as it was coined. They say: “what is used as it was convened upon in communication.”
And “figurative” is that which exceeds what it was coined as.
The use of mawḍūʿ (“coined”) in conjunction with the notion of iṣṭilāḥ (“convention”), and the opposition to “exceeding what it was coined as” (see quotation above) point to a conceptual understanding of waḍʿ: al-Juwaynī views it as the coinage of a word, from which its actual usage may differ—in this case, its figurative use (majāz).15On the term majāz in fourth/tenth-century theory, see Heinrichs, “On the Genesis of the ḥaqîqa-majâz Dichotomy.” Al-Juwaynī thus distills the debate down to this distinction.
Waḍʿ as a Term of Epistemology: dalīl waḍʿī and dalīl ʿaqlī
In addition to describing the status of individual words, waḍʿ also has a broader application in legal theory. Bernard Weiss’s rendering of waḍʿ as the “givenness” of language stresses the conventional nature of signification: words, signs or sounds are agreed upon and subsequently taken as a given. This agreement also holds for the status of scripture as a whole. Abū Bakr al-Bāqillānī distinguishes two types of legal evidence (dalīl, pl. adilla) in his work on legal theory, al-Taqrīb wa-l-irshād:16Al-Bāqillānī, al-Taqrīb wa-l-irshād (ed. Abū Zunayd), 204–205.
اعلموا – وفقكم الله – أن الأدلة على ضربين: فضرب منها عقلي والضرب الثاني وضعي..
فأما ما يدل بطريق المواضعة على دلالته، فنحو دلالات الألفاظ والرموز والإشارات والكتابة والعقود وأمثال ذلك مما يدل بعد مواضعة أهله على دلالة، ولولا مواضعتهم لما دل، ولا يتم التواضع على دلالة شيء دون أن يكون مما يدرك، ويكون إلى العلم به طريق مراد، نحو الألفاظ والإشارات وغيرها من الأكوان وأفعال الجوارح. فأما الضمائر وأفعال القلوب التي لا طريق إلى العلم بحصوله، فلا تتم المواطأة على دلالتها.
Know—may God grant you success—that there are two types of evidence: one type is rational (ʿaqlī) and the second type is by agreement (waḍʿī) […]
As for that which signifies by way of establishment of its signification, this is like the signification of vocal forms and symbols and signs and writing and contracts and other examples of what signifies after its people’s establishment of signification; and had there been no establishment it would not signify, and there would be no establishment of signification of a thing without it being something that can be perceived, and there being a way of knowing it, such as the vocal forms and signs and other sorts of beings and actions of the limbs.
A dalīl waḍʿī is thus the type of evidence that is valid on account of the fact that it was agreed upon, established by its people (muwāḍaʿat ahlihi). The Qurʾan is valid legal evidence (dalīl), not by rational (syllogistic) means (ʿaqlī), but by agreement (waḍʿī).17See also Ibrahim, “Reason and Revelation in Fakhr al-Dīn al-Rāzī and the Ashʿarī Tradition.” Put differently, the words and letters of the Qurʾan can only be understood to signify something because the meaning of these signs was agreed upon. Our knowledge of this meaning is by transmission (samʿī) through kalām al-ʿarab.
In al-Inṣāf, al-Bāqillānī distinguishes three types of dalīl:18al-Bāqillānī, al-Inṣāf (ed. al-Kawthari), 15.
عقلي: له تعلق بمدلوله، نحو دلالة الفعل على فاعله، وما يجب كونه عليه من صفاته، نحو حياته وعلمه وقدرته وإرادته.
وسمعي شرعي: دال من طريق النطق بعد المواضعة، ومن جهة معنى مستخرج من النطق.
ولغوي: دال من جهة المواطأة والمواضعة على معاني الكلام، ودلالات الأسماء والصفات وسائر الألفاظ …
- Rational evidence: this is connected to its referent, like the verb signifying its agent, and what attributes it necessarily has, such as its being alive or knowing or able or willing.
- Transmitted evidence from Revelation: this signifies by way of speech after the establishment, and with respect to a meaning drawn from speech.
- Linguistic evidence: this signifies with respect to the agreement upon and establishment of the meanings of language, the signification of names and attributes and the rest of the vocal forms.
This threefold distinction seems to foreshadow the later development of ʿilm al-waḍʿ, the discipline treating the intrinsic functions of words, not their individual meanings. By differentiating between transmitted (samʿī sharʿī) and linguistic (lughawī) evidence, al-Bāqillānī suggests that the former must be taken as a given whereas the latter can be analyzed. A similar distinction is made by referring to lexicography as matn al-lugha (the “matter/corpus” of language, see below) while terming the analysis of the categories into which lexemes may fall—and the consequences this may have for legal interpretation—waḍʿ al-lugha.
Conclusion
The two manuscript pages of ʿAḍud al-Dīn al-Ījī’s treatise on al-waḍʿ do not offer a systematic exposition of a theory of waḍʿ, but his achievement was to teach—and thereby to dictate or put into writing—some of the categories in which language could be divided, as a “necessary preliminary to legal theory.”19Weiss, “Language in Orthodox Muslim Thought,” 90. Al-Ījī categorizes demonstrative pronouns, personal pronouns, and particles as al-waḍʿ kullī wa-l-mawḍūʿ lahu mushakhkhaṣ: “the coinage is universal and the coined (i.e., the signified) is individual.” Such words may formally and without context be considered “universals” (kulliyyāt) but in practice, they are particulars (juzʾiyyāt), e.g., the third-person pronoun huwa is considered to designate not the general notion of “a single male person,” but the specific (khāṣṣ) male person it refers to in the context of speech.
In his commentary on the Risāla, al-Sharīf al-Jurjānī systematizes al-Ījī’s concise and elliptical wording, formulating all categories that follow logically from the Risāla: al-waḍʿ al-ʿāmm li-mawḍūʿ lahu ʿāmm, “general coinage for a general signified” and al-waḍʿ al-ʿāmm li-mawḍūʿ lahu khāṣṣ, “general coinage for a specific signified”—note that the terms ʿāmm and khāṣṣ have replaced kullī and mushakhkhaṣ. Subsequent commentators aimed to fit all word classes with their specific properties into this framework, yielding a fully-fledged discipline with its own categorizations and debates.20See Weiss, “Language in Orthodox Muslim Thought,” 90ff, for a detailed discussion of the theory and the ascription of certain elements to certain commentators.
The brief discussion of waḍʿ shows how a broad linguistic concept was made productive for systematic inquiry across several fields of scholarship. While the question of the origin of language was eventually considered irrelevant (as evidenced by the suspension of a final judgement on the matter by Ibn Jinnī), the concept of waḍʿ was used in legal theory to discuss the relationship between literal and figurative speech and to describe the epistemological status of scripture. The establishment of the rules of waḍʿ in treatises and commentaries from the eighth/fourteenth century onward and the gradual emergence of ʿilm al-waḍʿ also had conceptual and terminological consequences for the discipline of language (ʿilm al-lugha), which was considered a traditional (naqlī) discipline, not a rational science (ʿaqlī). When waḍʿ al-lugha was integrated into the curriculum, ʿilm al-lugha, the collection and verification of individual words, gathered through samāʿ and corroborated through authoritative quotations (shawāhid), was henceforth sometimes termed matn al-lugha, e.g., by the Ottoman scholars Mollā Fenārī (d. 834/1431),21Mollā Fenārī, Anmūdhaj al-ʿulūm, available, for instance, in the MS Süleymaniye Yazma Eser Kütüphanesi, Fatih 3677. and Sācaḳlızāde Meḥmed Efendi (d. 1145/1732). The latter defines ʿilm al-lugha as “the discipline of the individual coinage of single words” (ʿilm al-awḍāʿ al-shakhṣiyya li-l-mufradāt), thus fully integrating the terminology of ʿilm al-waḍʿ into a definition of its neighbouring (and predecessor) discipline.22ʿilm al-awḍāʿ al-shakhṣiyya li-l-mufradāt, Saçaklızāde, Tartīb al-ʿulūm (ed. al-Sayyid Ahmad), 119.
